4 replies to this topic

[Travesh]

Hi,

I am designing a reflux drum for a crude distillation unit. Operating parameters as below:
The reflux drum will separate the liquid water, liquid naptha and non-condensable vapor. The reflux drum pressure is controlled at 1.43 bar absolute. The overhead vapor is totally condensed by a shell & tube and an air-cooler in series. The amounts of liquid naptha and water can be obtained from my simulation however this is not the case for the non-condensables. Please advise how one can estimate the amount of non-condensable vapor. I propose 1% of the total liquid feed into the drum. Is this within typical industrial practice?


Crude tower: Top pressure = 1.77 bar absolute
Top temperature = 147 degC

Reflux drum: Pressure = 1.43 bar absolute
temperature = 87 degC

[Zauberberg]

First, you need to know the quantity of non-condensables (C1, C2) that come with the crude oil. I would suggest one simple method with reasonable accuracy, and that is to introduce tower overhead stream (at final condenser temperature and pressure) into a 2-phase separator and watch for vapor flow. Certain quantity of non-condensables will be dissolved in the wild naphtha stream, and rest will remain as vapor. Also, the presence of non-condensables will "strip" some quantity of C3+ from liquid hydrocarbon phase. Try this simple test and let us know the results - but first find the composition of crude oil.

[Travesh]

Dear Zauberberg,

Thank you kindly for your response.

Part of the difficulty I am experiencing is that the light ends portion of the crude feed does not contain any C1's or C2's. The composition (% by volume of crude) of light ends as follows:

Ethane: 0
Propane: 0.4
Iso-butane: 0.64
N-butane: 2.3
Iso-pentane: 1
N-pentane: 0.97

Therefore, from the simulation results, as recommended by you, on simulating the tower overhead fed into a two-phase separator (at the condenser temperature and pressure) , I obtain no vapor which implies total condensation.

Does these results imply a deviation from the ideal simulation for operating refineries since some proportion of non-condensables will always be present?

[Zauberberg]

If that is what you have got, that's it. For proper design, perhaps you could include these parameters as well:

1. Maximum expected condenser outlet temperature (e.g. during summer months) and see how things change
2. Changing the crude type (more non-condensibles?) in future

Also, an interesting thing is how the tower pressure will be controlled. You need to provide the loop that will maintain column pressure in cases when there is no vapor flow from outlet of the final condenser. Look at some sort of split-range control but be extremely careful when choosing the source of make-up gas: if there is a debutanizer tower downstream of the overhead receiver that produces pure C3/C4 mixture, connecting overhead receiver on the crude tower with fuel gas system will introduce C2- in the system and very likely contaminate your LPG product.

[Travesh]

Dear Zauberberg,

Thank you kindly for your comments. These were of great assistance.

Best regards,